Tag Archives: Predictive modelling

Type: Evidence | Proposition: A: Learning | Polarity: | Sector: | Country:


This empirical study investigates students' learning choices for mathematics and statistics in a blended learning environment, composed of both online and face-to-face learning components. The students (N = 730) were university freshmen with a strong diversity in prior schooling and a wide range of proficiency in quantitative subjects. In this context, we investigated the impact that individual differences in achievement emotions (enjoyment, anxiety, boredom, hopelessness) had on students' learning choices, in terms of the intensity of using the online learning mode versus the face-to-face mode. Unlike the general level of learning activities, which is only minimally influenced by achievement emotions, these emotions appear to have a moderately strong effect on a student's preference for online learning. Following this, we explored the antecedents of achievement emotions. Through the use of path-modeling, we conclude that while goal setting behavior only marginally impacts achievement emotions, effort views—a crucial component of the social-cognitive model of implicit theories of intelligence—have a substantial impact on achievement emotions.

Citation: Tempelaar, D. T., Niculescu, A., Rienties, B., Giesbers, B., & Gijselaers, W. H. (2012). How achievement emotions impact students' decisions for online learning, and what precedes those emotions. Internet and Higher Education, 15 (3), 161-169. | Url: http://www.sciencedirect.com/science/article/pii/S1096751611000704

Type: Evidence | Proposition: B: Teaching | Polarity: | Sector: | Country:

This paper focuses on work conducted at The Open University (OU), one of the world’s largest distance learning institutions, into predicting students who are at risk of failing a module. Since tutors at the university do not interact face to face with students, it can be difficult for to identify and respond to students who are struggling in time to resolve the difficulty. Predictive models were developed and tested using historic Virtual Learning Environment (VLE) activity data, combined with other data sources, for three OU modules. These revealed that it is possible to predict student failure by looking for changes in user activity in the VLE. More focused analysis of these modules yielded some promising results for the creation of accurate hypotheses about students who fail.

Comment by Martyn Cooper: This paper reports an investigation in online tutoring of the relation between learning analytic metrics and retention. It highlights the important result that little can be deduced from absolute levels of interaction but changes in level of interaction are significant. Although a limited study in the context of one VLE and a few modules it is likely to be a paper with wider significance.

Citation: A. Wolff, Z. Zdrahal, A. Nikolov and M. Pantucek, (2013). Improving retention: predicting at-risk students by analysing clicking behaviour in a virtual learning environment, In: Third International Learning Analytics and Knowledge Conference (LAK13), 8-12 April, Leuven, Belgium. | Url: http://dl.acm.org/citation.cfm?id=2460296